The phase III DUO-E study evaluated the addition of the monoclonal antibody durvalumab to chemotherapy, and the benefit of maintenance durvalumab, with and without the PARP inhibitor olaparib, in advanced endometrial cancer. As reported this past year,1 the inclusion of durvalumab reduced the risk of disease progression by 29%; the further addition of olaparib reduced the risk by 45% in the intent-to-treat population.
Kathleen Moore, MD, MS
At the 2025 Society of Gynecologic Oncology (SGO) Annual Meeting on Women’s Cancer, investigators reported results of a deeper dive into the robust impact of durvalumab plus olaparib on the mismatch repair–proficient (pMMR) population—which accounted for 80% of the study group.2 “We found that the heterogeneity of the molecular biomarkers in this population is striking, as is the amount of overlap of molecular biomarkers,” said Kathleen Moore, MD, MS, the Virginia Kerley Cade Endowed Chair in Cancer Development and Professor of Gynecologic Oncology at the University of Oklahoma Health Stephenson Cancer Center, Oklahoma City.
“The progression-free survival benefit provided by the addition of olaparib maintenance was observed across a range of biomarker and histological subgroups and in patients with detectable circulating tumor DNA (ctDNA) at baseline,” she said.
About DUO-E
The global phase III DUO-E/GOG-3041/ENGOT-EN10 trial enrolled 718 patients with newly stage III or IV recurrent endometrial cancer, naive to first-line systemic treatment of advanced disease and naive to PARP inhibitors and immune mediators. The primary endpoints were investigator-assessed progression-free survival for the durvalumab arm vs controls and durvalumab/olaparib arm vs controls.
All patients received up to six cycles of carboplatin and paclitaxel in combination with placebo or durvalumab, with responders transitioned to the maintenance phase. The arms follow:
- Control: Chemotherapy plus placebo followed by placebo maintenance
- Durvalumab: Chemotherapy plus durvalumab followed by maintenance durvalumab plus olaparib placebo
- Durvalumab/Olaparib: Chemotherapy plus durvalumab followed by maintenance durvalumab plus olaparib.
As reported previously,1 DUO-E met its dual primary endpoints, showing improvements in progression-free survival in the chemotherapy/durvalumab arm vs chemotherapy (hazard ratio [HR] = 0.71; P = .003) and chemotherapy/durvalumab plus olaparib maintenance arm vs chemotherapy (HR = 0.55; P < .0001) in the intent-to-treat population. The greatest benefit with chemotherapy and durvalumab was in the MMR-deficient (dMMR) subpopulation, where the median progression-free survival was not reached with durvalumab (HR = 41) and was 31.8 months in the durvalumab/olaparib arm (HR = 0.41), vs chemotherapy alone.
“The outcome in the dMMR subgroup was consistent with other randomized phase III studies. The addition of an immune checkpoint inhibitor, durvalumab, was transformational in this patient population,” Dr. Moore said. In contrast, olaparib did not enhance the benefit in the dMMR subset, “but that changed in the pMMR subset, where it did,” she said. “This was very exciting, but begged the next question: Does the addition of olaparib benefit all patients with pMMR endometrial cancer, across all the increasingly recognized molecular subsets?”
Histologic Heterogeneous pMMR Subset
Dr. Moore presented post hoc exploratory analyses of relevant clinical markers to assess biomarker and histologic heterogeneity, presence of baseline ctDNA, and efficacy of these treatments in the pMMR subpopulation of 575 patients evaluable for all markers. The analysis found this group to be highly heterogeneous, with multiple genomic drivers and frequent overlaps of biomarkers and histology. In the biomarker-known subpopulation, 84% of patients tested positive for at least one biomarker, and among patients evaluated for ctDNA, 79% had detectable levels at baseline. For 289 patients, baseline ctDNA could be analyzed. “All the molecular subsets had very high detection of ctDNA,” she reported.
In the tumor analysis, PD-L1 (67%) and TP53 mutations (59%) were the most prevalent biomarkers, and they overlapped in 44% of patients. Other common biomarkers included homologous recombination repair mutations (21%), BRCA mutations (8%), and POLE mutations (2%); serous histology was observed (27%), and these tumors largely overlapped with TP53-mutated tumors.
KEY POINTS
- An exploratory post hoc analysis of DUO-E showed the mismatch repair–proficient (pMMR) subpopulation to be highly heterogeneous, with frequent overlap of biomarkers and histology as well as a high prevalence of detectable baseline circulating tumor DNA (ctDNA).
- Despite the heterogeneity, the addition of olaparib maintenance to durvalumab and chemotherapy enhanced the progression-free survival benefit offered by durvalumab, vs carboplatin/paclitaxel alone. Post hoc, exploratory analysis suggested benefit across a range of biomarker and histologic subsets, including patients with detectable baseline ctDNA.
Durvalumab’s benefit was relatively consistent across all biomarkers, as was also true for durvalumab plus olaparib. “The benefit of adding olaparib in this population is not driven by one particular molecular subset,” Dr. Moore added.
Dr. Moore said the “most interesting hypothesis-generating point” was that the benefit of olaparib was particularly striking in patients with detectable ctDNA. In this subset of patients, the hazard ratio for chemotherapy plus durvalumab (without olaparib) vs chemotherapy was 0.61; the addition of olaparib boosted this hazard ratio to 0.36. “Does that signify that ctDNA at baseline is a predictive biomarker for adding olaparib? No, not at this point. It may be something to tease out later with more data,” Dr. Moore maintained.
DISCLOSURE: Dr. Moore reported financial relationships with Genentech/Roche, Immunogen, AstraZeneca, VBL Therapeutics, Merck, Eisai, Myriad Genetics, OncXerna Therapeutics, Onconova Therapeutics, Mereo BioPharma, Novartis, Verastem/Pharmacyclics, Aadi, Clovis Oncology, Caris Life Sciences, Hengrui Pharmaceutical, Novartis/Pfizer, Iovance Biotherapeutics, Janssen Oncology, Regeneron, Zentalis, and GlaxoSmithKline.
REFERENCES
1. Westin SN, et al: J Clin Oncol 42:283-299, 2024.
2. Moore K, et al: 2025 SGO Annual Meeting on Women’s Cancer. Abstract. Presented March 15, 2025.
EXPERT POINT OF VIEW
The post hoc analysis of DUO-E is seeking to identify which biomarkers might guide the selection of chemotherapy and immunotherapy, vs chemotherapy/immunotherapy plus a PARP inhibitor, within the endometrial cancer population with mismatch repair–proficient (pMMR) tumors, said the study’s invited discussant Ramez N. Eskander, MD, the Clinical Trials Office Director at the University of California San Diego Health, Rebecca and John Moores NCI Designated Comprehensive Cancer Center, San Diego. At this point, there remain more questions than answers, he concluded.
Ramez N. Eskander, MD
DUO-E is the first clinical trial to report on the potential relevance of circulating tumor DNA (ctDNA) on clinical efficacy, he noted. In general, the highest ctDNA detection rates were in patients whose molecular profile was unknown; therefore, the concordance between ctDNA and genomic assessment cannot yet be accurately determined. The molecular-unknown cohort was enriched with patients enrolled from China, and whether this may have any relevance is also unknown.
Furthermore, Dr. Eskander added, it is “difficult to interpret” the clinical relevance of the different molecular subsets identified in the analysis, primarily because the various studies in endometrial cancer have produced “mixed messages,” he said. “We’ve had data from DUO-E,1 NRG GY018,2 and GOG-3031/RUBY3 that suggest differential efficacy signals based on the biomarker of interest…. And these discordant findings make us wonder where the truth lies,” Dr. Eskander commented.
He emphasized that Forest plots showing the magnitude of benefit of olaparib among the molecular subsets of patients must be recognized as exploratory. “These subsets are not independent; and patients may have multiple biomarkers simultaneously. Interest remains in identifying relevant biomarkers in the pMMR population. We must be thoughtful about multiple comparisons. The only way to confirm these exploratory, hypothesis-generating findings is through the design and conduct of prospective trials.”
DISCLOSURE: Dr. Eskander has served as a consultant or advisor for AstraZeneca, Clovis Oncology, Daiichi Sankyo, Eisai, Elevar Therapeutics, MSD, GlaxoSmithKline, ImmunoGen, Mersana Therapeutics, Myriad Genetics, Novocure GmbH, Onconova Therapeutics, Nuvectis, PMV Pharmaceuticals, Regeneron, Lilly, AbbVie, and Pfizer.
REFERENCES
1. Moore K, et al: 2025 SGO Annual Meeting on Women’s Cancer. Abstract. Presented March 15, 2025.
2. Eskander RN, et al: N Engl J Med 388:2159-2170, 2023.
3. Powell MA, et al: Ann Oncol 35:728-738, 2024.
